人工智能在年龄相关性黄斑变性领域的研究进展

近年来,人工智能（AI）在医学领域呈现出迅猛发展的态势,在眼科学的应用范围也不断扩大。年龄相关性黄斑变性（AMD）是一种常见致盲性眼病,AI在AMD中的研究与应用日益增多,加深了人们对AMD的认识。利用AI技术对AMD患者进行早期筛查与诊断可以降低患者视功能损伤的风险,其次运用AI对AMD进展进行预测,设计个性化治疗方案,可以改善患者预后。该文综述了AI在AMD筛查、诊断、预后中的最新进展,以及在临床实践上面临的困难与挑战,旨在为将来的研究提供新思路。     

Framingham高血压发病风险预测模型对新疆哈萨克族牧民发生高血压的预测价值研究

背景　新疆哈萨克族牧民高血压患病率高,寻找有效的方法对哈萨克族牧民高血压的发病风险进行评估,并进行合理的干预,预防高血压事件的发生十分重要。目的　评价Framingham高血压发病风险模型（以下简称Framingham模型）预测新疆哈萨克族牧民发生高血压的准确性和适用性,为建立适用于新疆哈萨克族牧民高血压发病风险预测模型提供依据。方法　2008年1月,采用分层整群随机抽样法,以新疆乌鲁木齐县南山牧区的5 327例哈萨克族牧民为研究对象,建立南山动态队列。收集其基线资料（包括问卷调查表和体格检查表）,并进行2年1次的随访调查,共3次。随访截至2018年11月,结局事件为发生高血压。将研究对象随机分为建模队列（60%哈萨克族牧民,3 196例）和验证队列（40%哈萨克族牧民,2 131例）,按照Framingham模型相同的方法及预测因素,采用多因素weibull回归分析、依据建模队列调整Framingham模型,即为依据建模队列调整后的Framingham模型。采用区分能力和标定能力验证上述依据本研究对象数据调整后的Framingham模型对验证队列高血压发病风险进行预测。结果　截至2018年11月,1 985例哈萨克族牧民发生高血压。累计共随访16 897人年,高血压发病率为11.75/100人年〔95%CI（11.27/100人年,12.24/100人年）〕。本研究验证队列人群经2年、4年随访,分别有269例、562例发生高血压。依据建模队列调整后的Framingham模型与原Framingham模型预测随访2年验证队列人群发生高血压的受试者工作特征曲线下面积（AUC）分别为：AUC依据建模队列调整后=0.647〔95%CI（0.624,0.670）〕和AUC原Framingham模型=0.594〔95%CI（0.571,0.617）〕;依据建模队列调整后的Framingham模型与原Framingham模型预测随访2年验证队列人群发生高血压的AUC比较,差异有统计学意义（χ2=5.085,P<0.05）;依据建模队列调整后的Framingham模型与原Framingham模型预测随访4年验证队列人群发生高血压的AUC分别为：AUC依据建模队列调整后=0.609〔95%CI（0.590,0.628）〕和AUC 原Framingham模型=0.588〔95%CI（0.569,0.607）〕;依据建模队列调整后的Framingham模型与原Framingham模型预测随访4年验证队列人群发生高血压的AUC比较,差异有统计学意义（χ2=3.448,P<0.001）。依据建模队列调整后的Framingham模型与原Framingham模型预测随访2年验证队列人群高血压发病率与实际发病率比较Hosmer-Lemeshow χ2（H-L χ2）检验值分别为697.68（P<0.05）、802.40（P<0.05）;预测随访4年验证队列人群高血压发病率与实际发病率比较,H-L χ2检验值分别为682.61（P<0.05）、832.82（P<0.05）。结论　Framingham模型在预测新疆哈萨克族牧民高血压的发病风险中区分能力和标定能力均较差,不能很好地预测新疆哈萨克族牧民高血压的发病风险,需要构建更适合该人群的高血压发病风险预测模型。     

Predicting in-hospital mortality in ICU patients with sepsis using gradient boosting decision tree

Sepsis is a leading cause of mortality in the intensive care unit. Early prediction of sepsis can reduce the overall mortality rate and cost of sepsis treatment. Some studies have predicted mortality and development of sepsis using machine learning models. However, there is a gap between the creation of different machine learning algorithms and their implementation in clinical practice.This study utilized data from the Medical Information Mart for Intensive Care III. We established and compared the gradient boosting decision tree (GBDT), logistic regression (LR), k-nearest neighbor (KNN), random forest (RF), and support vector machine (SVM).A total of 3937 sepsis patients were included, with 34.3% mortality in the Medical Information Mart for Intensive Care III group. In our comparison of 5 machine learning models (GBDT, LR, KNN, RF, and SVM), the GBDT model showed the best performance with the highest area under the receiver operating characteristic curve (0.992), recall (94.8%), accuracy (95.4%), and F1 score (0.933). The RF, SVM, and KNN models showed better performance (area under the receiver operating characteristic curve: 0.980, 0.898, and 0.877, respectively) than the LR (0.876).The GBDT model showed better performance than other machine learning models (LR, KNN, RF, and SVM) in predicting the mortality of patients with sepsis in the intensive care unit. This could be used to develop a clinical decision support system in the future.     

The role of metabolomic markers for patients with infectious diseases: implications for risk stratification and therapeutic modulation

Introduction: Metabolomics is a rapidly growing area of research. Metabolomic markers can provide information about the interaction of different organ systems, and thereby improve the understanding of physio-pathological processes, disease risk, prognosis and therapy responsiveness in a variety of diseases.

Areas covered: In this narrative review of recent clinical studies investigating metabolomic markers in adult patients presenting with acute infectious disease, we mainly focused on patients with sepsis and lower respiratory tract infections. Currently, there is a growing body of literature showing that single metabolites from distinct metabolic pathways, as well as more complex metabolomic signatures are associated with disease severity and outcome in patients with systemic infections. These pathways include, among others, metabolomic markers of oxidative stress, steroid hormone and amino acid pathways, and nutritional markers.

Expert commentary: Metabolic profiling has great potential to optimize patient management, to provide new targets for individual therapy and thereby improve survival of patients. At this stage, research mainly focused on the identification of new predictive signatures and less on metabolic determinants to predict treatment response. The transition from observational studies to implementation of novel markers into clinical practice is the next crucial step to prove the usefulness of metabolomic markers in patient care.     

Independent validation of the ICU requirement score in a cohort of acutely poisoned adults

Objective: To independently validate the predictive value of the intensive care requirement score (IRS) in unselected poisoned patients.

Design: Retrospective chart review.

Patients and methods: Five hundred and seventeen out of 585 admissions for acute intoxications could be analyzed. Eleven were excluded for a condition already requiring intensive care unit (ICU) support at admission (e.g., preclinical intubation). A further 57 admissions were excluded due to missing data. The IRS was calculated using a point-scoring system including age, Glasgow Coma Scale, heart rate, type of intoxication, and preexisting conditions. It was then compared to a composite endpoint indicating an ICU requirement (death in hospital, vasopressors, need for ventilation). The endpoint and the point-scoring system were identical to the original publication of the score.

Results and conclusion: Twenty-three out of 517 patients had a complicated clinical course as defined by meeting the endpoint definition. Twenty-one out of 23 complicated courses had a positive IRS (defined as greater or equal 6 points), as compared to 255/494 patients with an uncomplicated clinical course (p?相似文献   

Real-time toxicity prediction of Aconitum stewing system using extractive electrospray ionization mass spectrometry

Due to numerous obstacles such as complex matrices, real-time monitoring of complex reaction systems (e.g., medicinal herb stewing system) has always been a challenge though great values for safe and rational use of drugs. Herein, facilitated by the potential ability on the tolerance of complex matrices of extractive electrospray ionization mass spectrometry, a device was established to realize continuous sampling and real-time quantitative analysis of herb stewing system for the first time. A complete analytical strategy, including data acquisition, data mining, and data evaluation was proposed and implemented with overcoming the usual difficulties in real-time mass spectrometry quantification. The complex Fuzi (the lateral root of Aconitum)–meat stewing systems were real-timely monitored in 150 min by qualitative and quantitative analysis of the nine key alkaloids accurately. The results showed that the strategy worked perfectly and the toxicity of the systems were evaluated and predicated accordingly. Stewing with trotters effectively accelerated the detoxification of Fuzi soup and reduced the overall toxicity to 68%, which was recommended to be used practically for treating rheumatic arthritis and enhancing immunity. The established strategy was versatile, simple, and accurate, which would have a wide application prospect in real-time analysis and evaluation of various complex reaction systems.